The Book

Quarto

Quarto enables you to weave together content and executable code into a finished document. To learn more about Quarto see https://quarto.org.

The Book

By Sam Assaf

Outline

• Abstract & Example : Analysts & Coaches
• Introduction : Analysts & Coaches
• Problem Framing : Analysts
• Related works : Analysts
• Data Overview : Analysts
  • Variables in Model : Analysts & Coaches
  • Data Visualization : Analysts & Coaches
• Methods : Analysts
  • Model Adjustments : Analysts & Coaches
    • Model Without Risk Aversion Adjustments : Analysts & Coaches
    • Model With Risk Aversion Adjustments : Analysts & Coaches
• Discussion : Analysts & Coaches
• Examples Plots / Situation : Analysts & Coaches
• Conclusion and Future Work : Analysts & Coaches

Abstract & Example : Analysts & Coaches

FILL IN CAPTION HERE.

FILL IN CAPTION HERE.

Introduction : Analysts & Coaches

Problem Framing : Analysts

Related works : Analysts

Data Overview : Analysts

FILL IN CAPTION HERE.

Variables in Model : Analysts & Coaches

Outcome Variable	Game Data	Play Data
Win Probability Added (WPA)	Stadium Roof Surface Temperature Wind Weather detail	Play type (input) | Field goal probability | Yards to go (first down or | score) | Yardline (continuous and | binned)1 | EPAs2 | ELOs3 | Score differential (binned)4| Game time remaining (binned)5 Score difference time ratio6|

# Create vectors for each category of information
OutcomeVariable <- c("Win Probability Added (WPA)", '', '', '', '', '', '', '', '')
GameData <- c("Stadium", "Roof", "Surface", "Temperature", "Wind", "Weather detail", '', '', '')
PlayData <- c("Play type (input)", "Field goal probability", "Yards to go (first down or score)", 
              "Yardline (continuous and binned)³", "EPAs⁴", "ELOs⁵", "Score differential (binned)⁶", 
              "Game time remaining (binned)⁷", "Score difference time ratio⁸")

# Combine into a data frame
df <- data.frame(OutcomeVariable, GameData, PlayData, stringsAsFactors = FALSE)

# Create the gt table
gt_table <- df %>%
  gt() %>%
  cols_label(
    OutcomeVariable = "Outcome Variable",
    GameData = "Game Data",
    PlayData = "Play Data")
  # ) %>%
  # tab_spanner(label = "Game Data", columns = vars(GameData)) %>%
  # tab_spanner(label = "Play Data", columns = vars(PlayData))

# Print the gt table
gt_table

Outcome Variable	Game Data	Play Data
Win Probability Added (WPA)	Stadium	Play type (input)
	Roof	Field goal probability
	Surface	Yards to go (first down or score)
	Temperature	Yardline (continuous and binned)³
	Wind	EPAs⁴
	Weather detail	ELOs⁵
		Score differential (binned)⁶
		Game time remaining (binned)⁷
		Score difference time ratio⁸

Data Visualization : Analysts & Coaches

FILL IN CAPTION HERE.

FILL IN CAPTION HERE.

Methods : Analysts

FILL IN CAPTION HERE.

FILL IN CAPTION HERE.

Model Adjustments : Analysts & Coaches

Model Without Risk Aversion Adjustments : Analysts & Coaches

FILL IN CAPTION HERE.

Model With Risk Aversion Adjustments : Analysts & Coaches

FILL IN CAPTION HERE.

Discussion : Analysts & Coaches

Examples Plots / Situation : Analysts & Coaches

Conclusion and Future Work : Analysts & Coaches

Footnotes

1. Quantitative : 1 - 60 & Categorical / Bins : [1, 5], [6, 15], [16, 25], [26, 38], [39 to 45], [46, 60]

2. Expected Points Added (EPAs) are a measure of the expected point value at the start of a play. Various aggregated and lagged measures - for both teams, offensive pass & run game, defensive pass & run game.

3. ELO ratings are used to rank teams and specific features of a team based on their performance. Various aggregated and lagged measures - for both teams, offensive pass & run game, defensive pass & run game

4. Categorical / Bins : (-∞,17], [-16,-9], [-8,-4], [-3,0], [1,3], [4, 8], [9, 16], [17,∞)

5. Categorical / Bins : First Quarter : [15,0], Second Quarter : [15,4), [4, 2), [2,0], Third Quarter: [15,0], Fourth Quarter : [15, 7), [7, 4), [4, 2), [2, 0]

6. Score difference time ratio is addressing the relationship between score differential and time remaining. The average time remaining for each bin of time remaining and most frequently occurring score differential for each bin to calculate this measure